Surface Induced Conformational Changes of a Hard Protein Lysozyme

Changes in Solvation due to Adsorption

As illustrated in Fig. 3, a first extremely rapid isotopic exchange occurs within the first 10 min of incubation of lysozyme in the deuterated buffer, corresponding to the exchange in COND of 70% of the amino acid residues. The residual NH content of lysozyme after 2 h of incubation reaches 21% in pure phosphate buffer (Figs. 3A,C) and 16% in phosphate-buffered saline (Fig. 3B). The ionic strength is known to influence the NH/ND isotope exchange kinetics of proteins (Gregory and Lumry 1985).

The adsorption of lysozyme with a positive net charge on the pure silica surface is driven mainly by electrostatic interactions (Robeson and Tilton 1996; Wahlgren et al. 1993). The adsorbed amounts of lysozyme are increased at low ionic strength, as illustrated in Fig. 3A compared to Fig. 3B. The first molecules of lysozyme adsorbed on the negatively charged silica do not change their solvation state since their residual NH content is rather unchanged compared to the equilibrium NH content of the protein in solution. The effect of ionic strength is to limit both the adsorbed amount of lysozyme and the NH/ND exchange of the adsorbed molecules (Fig. 2B). Since the diffusion of D2O inside the adsorbed molecules of lysozyme is limited, no increase in hydrated random domains is observed (Figs. 4A,B). The adsorption of the positively charged lysozyme on the silica surface does not induce a significant change in either solvation or in secondary structure (Figs. 4A,B). The more exchanged states of adsorbed lysozyme in phosphate buffer appear when the adsorption density r exceeds the limit value of 1.25 mg m-2. This limitvalue of adsorption densitycould correspond to the adsorption of lysozyme in a side-on orientation, as reported by other groups (Claesson et al. 1995; Wahlgren et al. 1995). The fast adsorption process is considered to form a basal monolayer of lysozyme in native structure, and the slower process could correspond to the multilayer growth. In the second step, the subsequent adsorbed molecules of lysozyme would be more solvated than in the former monolayer (see Fig. 3A), although they did not undergo a conformational change on the negatively charged silica. Ball and

40 T

30-

20-

10-

40-

-»—*

30-

c

(1)

c

20-

8

X

10-

40-

30-

20-

10-

0-

0

solution

Lm.B'H-■--■

* 5-§

-5-

^--5-§

60 120 180 240

2.0

1.5

1.0

0.5

2.0

1.5

1.0

t

0.5

2.0

1.5

1.0

0.5

0.0

Time (min)

Fig. 3. Solvation changes (-0-) and adsorbed amount (-■-) of lysozyme adsorbed at a concentration of 250 |g ml-1 and pD 7.5 at different liquid-solid interfaces: a deuterated phosphate buffer-pure silica surface (A), deuterated phosphate buffer with 10 mM NaCl-pure silica surface (B) and deuterated phosphate buffer-CH3-terminated SAMs (C). The residual NH content (expressed as a percentage of the overall peptide carbonyls) is monitored by the amide II:amide I' ratio measurements. The adsorbed amount (r) of lysozyme is calculated from the adsorption density equation (Sperline et al. 1987)

Ramsden (2000) showed that lysozyme exhibits also a two-step adsorption kinetics on a hydrophilic silica-titania surface. Surface apparatus studies of lysozyme adsorbed onto mica suggest that lysozyme could exist as a dimer forming a partial bilayer (Blomberg et al. 1998). Su et al. (1998b) showed that the adsorbed lysozyme on silica retains its tertiary structure and that no significant denaturation occurs. The results of their neutron reflection studies suggest that the structural arrangement of the adsorbed lysozyme at the aqueous-silica interface is determined by the lateral electrostatic repulsion, which in turn is dependent on the jamming within the bilayers and the net charge of the protein at a given pH. From x-ray data, the basic and acidic residues of lysozyme are known to have an uneven distribution at its periphery (Wilson et al. 1992). A preferred orientation at low surface coverage would place the external basic residues of lysozyme facing the negatively charged silica surface. Consequently, this preferred orientation decreases the electrostatic repulsion between adjacent protein molecules and maximises the surface attraction, as already suggested by Haggerty and Lenhoff (1993b).

Many studies have underlined the orientation effect, which could occur during adsorption. When the bulk concentration of protein is low, a side-on orientation of the adsorbed molecules could be favoured. Rather, adsorption at a higher bulk concentration could lead to an end-on configuration due to a stronger protein-protein interaction, yielding to a more compact adsorbed layer. Obviously, the presented simplistic calculation from the adsorbed amount does not take into account that in some cases, the jamming coverage could be less than unity (Arai and Norde 1990) and the possibility of coexistence of different orientations of the adsorbed proteins at the interface (Su et al. 1998b). For instance, the experimental determinations of the thickness or the lateral distribution by other techniques is necessary to determine whether the formation of an incomplete monolayer or a loosely packed monolayer has occurred within side-on or end-on orientations.

Adsorption Kinetics and Conformational Changes

For adsorption on the CH3-terminated SAMs, the adsorption plateau is rapidly reached within 30 min, meanwhile the residual NH content decreases drastically to a plateau value of 5%. As for BSA adsorbed onto a hydrophobic support, some D2Omoleculespenetrateinsidethelysozymecore and solvate more amide CONH bonds because of the changes in secondary structure induced by adsorption (Noinville et al. 2002). However, the major conformational change concerning lysozyme adsorbed onto a hydrophobic surface consists in an a-helix to ^-sheet structural conversion (Noinville et al. 2002; Yokoyama et al. 2003). The maximum adsorbed amount of lysozyme of 0.8 mg m-2 corresponds to the close packing molecules with a cross-sectional area of 29 nm2 (Fig. 3C). The increase in the cross-sectional area of adsorbed lysozyme could be a consequence of the conformational changes of the molecules requiring a larger area of contact with the hy-drophobic surface. The losses of 12% of peptide carbonyls in helical structure and of 5% in random hydrated domains give rise to 8% of CO in ^3-structured domains and 7% of CO in self-associated domains (Noinville et al. 2002). Since one part of the newly formed domains is ^-sheeted, the molecular shape of adsorbed molecules of lysozyme is expected to be less spread than the soft BSA. The lower content in random hydrated domains of the adsorbed lysozyme is linked to the dehydration of the protein-hydrophobic support (Fig. 4C). The released peptide carbonyls natively involved in the helical structure of lysozyme could not form hydrogen bonds

Fig. 4. Comparison of secondary structures of lysozyme resulting from our spectral analysis in a deuterated solution and adsorbed at a bulk concentration of 250 |g ml-1 and pD 7.5 at different liquid-solid interfaces: a deuterated phosphate buffer-pure silica surface (A), deuterated phosphate buffered saline-pure silica surface (B) and deuterated phosphate buffer-CH3-terminated SAMs (C). The given percentage of amide I' values were recorded at 4 h in solution (clear bars) and for adsorption (filled bars), so that corresponds to the maximum adsorbed amount of lysozyme reported in Fig. 3

with water molecules since there are no more present at the dehydrated hydrophobic interface. The unfolded peptide carbonyls do form hydrogen bonds that are involved either in intramolecular domains (^-sheets) or in intermolecular self-associated domains. The global structural change in lysozyme caused by adsorption onto a hydrophobic support involves a small loss of stereoregular structures concerning 4% of the polypeptide backbone against 13% for the soft BSA (Noinville et al. 2002).

Was this article helpful?

0 0

Post a comment